A5101944838- Water Systems and Optimization
- Machine Fault Diagnosis Techniques
- Structural Integrity and Reliability Analysis
- Advanced Algorithms and Applications
- Anomaly Detection Techniques and Applications
- Infrastructure Maintenance and Monitoring
- Advanced Sensor and Control Systems
- Non-Destructive Testing Techniques
- Power Transformer Diagnostics and Insulation
- Structural Health Monitoring Techniques
- Fire Detection and Safety Systems
- High voltage insulation and dielectric phenomena
- Oil and Gas Production Techniques
- Cellular and Composite Structures
- Image and Signal Denoising Methods
- Image Processing and 3D Reconstruction
- Image Processing Techniques and Applications
- Geotechnical Engineering and Underground Structures
- Tunneling and Rock Mechanics
- Structural Response to Dynamic Loads
- Fault Detection and Control Systems
- Emotion and Mood Recognition
- Blind Source Separation Techniques
- Advanced Machining and Optimization Techniques
- Industrial Technology and Control Systems
Southwest Jiaotong University
2024-2025
China University of Geosciences (Beijing)
2025
Energy Research Institute
2021-2024
Northeast Petroleum University
2017-2024
Shenzhen Institutes of Advanced Technology
2024
Chinese Academy of Sciences
2024
Stomatology Hospital
2024
Education Department of Heilongjiang Province
2021-2024
Harbin Institute of Technology
2017-2024
Shanghai Children's Medical Center
2024
This paper considers the problem of effective feature extraction acoustic signals from oil and gas pipelines under different working conditions. A pipeline leakage detection method is proposed based on multi-feature entropy fusion local linear embedding (LLE). First, seven kinds commonly used which can reflect characteristics signal better are extracted through experiments, including permutation entropy, envelope approximate fuzzy energy sample dispersion entropy. The seven-dimensional...
When the gas pipeline leaks, it causes huge economic losses. This paper establishes a digital twin model of based on pressure signal generated by leak and researches detection. First, an online updating is established to update data physical information space parameters online. Second, visual display spatial pipelines output in real-time. If leakage identified, alarm would be triggered corresponding emergency rescue plan initiated leakage. Finally, identification can analysing finite element...
This paper is concerned with the development of improved multi-strategy MPA-VMD method and its application in pipeline leakage detection. Aiming at shortcomings marine predator algorithm (MPA) itself, which has a slow convergence speed easy to fall into local optimum, an MPA proposed used find two important parameters variational mode decomposition (VMD), then dynamic entropy select effective modes. In initial stage population, good point set strategy adopted enhance search accuracy by...
Traffic flow prediction is a very important research field in intelligent transportation system. The traditional methods have wide application traffic prediction. However, the short-term prediction, due to complexity of its influencing factors, cannot predict well. In this paper, model constructed by using and memory network, modal aliasing problem solved variational decomposition. From experimental results, method proposed paper suitable for can achieve good effect accuracy.
Large Multimodal Models (LMMs) exhibit major shortfalls when interpreting images and, by some measures, have poorer spatial cognition than small children or animals. Despite this, they attain high scores on many popular visual benchmarks, with headroom rapidly eroded an ongoing surge of model progress. To address there is a pressing need for difficult benchmarks that remain relevant longer. We take this idea to its limit introducing ZeroBench-a lightweight reasoning benchmark entirely...
Abstract In order to solve the problem of recognition error caused by noise interference in oil and gas pipeline signal traditional leakage detection relies on expert experience extract features manually, an model based deep learning is proposed this paper. The consists data preprocessing part pattern part. Firstly, a denoising algorithm variational mode decomposition (VMD) Manhattan distance (MD) reduce quality subsequent process. Secondly, combined with one-dimensional temporal...
In order to separate the effective components and noise after variational mode decomposition (VMD) improve denoising effect of VMD, a approach combining VMD with energy value (EV) (VMD-EV-VMD) is proposed. First, decomposes original signal into K intrinsic functions (IMF) then calculates EV probability density function each IMF component. According changing trend value, are distinguished. Subsequently, component decomposed by again, selected calculating correlation coefficient (CC) between...
In this paper, an image denoising algorithm is presented based on the two-dimensional variational mode decomposition (2D-VMD) and Hausdorff distance (HD). The procedure of developed that: (1) use 2D-VMD to decompose into a number intrinsic functions (IMFs); (2) HD probability density function (PDF) distinguish signal-dominant IMFs (S-D IMFs) noise-dominant (N-D IMFs), then wavelet threshold method eliminate noise in N-D IMFs; (3) denoised S-D reconstruct obtain image. effectiveness proposed...
This paper is concerned with the pipeline leakage detection problem. A signal feature extraction method based on improved variational mode decomposition (VMD) and multi-feature fusion are proposed. First of all, number K-value VMD determined by combining empirical (EMD) centre frequency method. Next, according to variance contribution rates, effective components selected from obtained VMD, subsequently, reconstructed get de-noised signal. Then, characteristic parameters that might...
With the continuous development of pipeline transportation industry, leakage often occurs, posing a great threat to people’s lives and property safety. In order improve detection accuracy natural gas leakage, method based on improved variational mode decomposition algorithm Lempel–Ziv complexity analysis is proposed. this work, normalized mutual information used determine level K decomposition, extract signal feature. The results show that proposed has higher classification than other...
Issues concerning natural gas pipeline leakage are becoming more prominent than ever because of the continuing expansion networks. Although many scholars have extensively investigated generation and detection methods for acoustic signals, systematic research on characteristics interference signals remains insufficient. Results show that method based RBF kernel function is feasible fault diagnosis, yielding 100% sensitivity, 92% specificity, 96% accuracy.